MMM Data Model Enhances Knowledge Interoperability for AI.

Many existing information systems are built around documents, which, while good for dissemination, restrict how knowledge can be structured, updated, and reused. While formal approaches exist, they often lack widespread adoption due to prioritizing strict structure over usability. With AI systems increasingly reshaping document production, there's a growing need for a portable, unified alternative for human knowledge expression and exchange. This research presents the MMM data model, developed from the practical needs of interdisciplinary collaborative research. MMM aims to provide a normative specification for knowledge interoperability in a decentralized knowledge commons. It achieves this by balancing a small set of normative constraints with the flexibility of free-text labels, making it adaptable across various disciplines, applications, and deployments without requiring semantic convergence. Early implementations and pilot data demonstrate its feasibility and usability.